An efficient ME architecture for high definition videos using the new MPDS algorithm

Authors:
Marcelo Schiavon Porto;Gustavo Sanchez;Diego Noble;Luciano Agostini;Sergio Bampi
Affiliations:
Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil;Universidade Federal de Pelotas, Pelotas, Brazil;Universidade Federal de Pelotas, Pelotas, Brazil;Universidade Federal de Pelotas, Pelotas, Brazil;Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil
Venue:
Proceedings of the 24th symposium on Integrated circuits and systems design
Year:
2011

Citing 6
Cited 1

A configurable motion estimation architecture for block-matching algorithms

IEEE Transactions on Circuits and Systems for Video Technology
A high-performance three-engine architecture for H.264/AVC fractional motion estimation

IEEE Transactions on Very Large Scale Integration (VLSI) Systems
Dynamically variable step search motion estimation algorithm and a dynamically reconfigurable hardware for its implementation

IEEE Transactions on Consumer Electronics
Hybrid parallel motion estimation architecture based on fast top-winners search algorithm

IEEE Transactions on Consumer Electronics
A new diamond search algorithm for fast block-matching motion estimation

IEEE Transactions on Image Processing
A Hardware-Efficient Multi-Resolution Block Matching Algorithm and its VLSI Architecture for High Definition MPEG-Like Video Encoders

IEEE Transactions on Circuits and Systems for Video Technology

DMPDS: a fast motion estimation algorithm targeting high resolution videos and its FPGA implementation

International Journal of Reconfigurable Computing

Quantified Score

Hi-index	0.01

Visualization

Abstract

This paper presents an efficient hardware architecture for motion estimation (ME) in high resolution digital videos. This architecture is built upon the new Multi-Point Diamond Search algorithm (MPDS) which is a fast algorithm that increases the ME quality when compared with other fast algorithms. The MPDS is able to reduce local minima falls, increasing the quality results of the motion vectors since less residual energy needs to be further encoded, especially in high definition videos. Two versions of this architecture are presented. The first one focuses on high performance, and the second one focuses on quality improvement. Both architectures were described in VHDL and synthesized to Stratix 4 FPGA and TSMC 90nm standard cells technology. The synthesis results show that both architectures achieve performance to process HD 1080p videos at 30 fps. These solutions also presented the lowest power consumption (4.5mW and 9mW) in comparison to the related works in the field, with a low hardware resources utilization.